The Agent Skills Directory

[SAFE]: The skill implement functionality using well-established and trusted scientific Python libraries.\n
Dependencies include scanpy, pandas, numpy, and matplotlib.\n
The implementation is consistent with the stated purpose of analyzing single-cell genomic data.\n- [EXTERNAL_DOWNLOADS]: No unauthorized or suspicious remote code downloads were detected.\n
The skill only references official documentation and community resources for the scanpy project and the scverse ecosystem.\n
No executable content or binaries are fetched from external servers.\n- [COMMAND_EXECUTION]: The provided Python scripts and command examples use safe practices for local data processing.\n
scripts/qc_analysis.py correctly uses argparse for command-line argument handling.\n
File system operations are limited to creating standard directories for results and figures using os.makedirs.\n- [DATA_EXFILTRATION]: There is no evidence of data exfiltration or unauthorized access to sensitive files.\n
The skill performs purely local analysis on user-provided datasets (e.g., .h5ad, .csv).\n
No network-enabled tools or libraries (like requests or socket) are used to transmit data externally.\n- [PROMPT_INJECTION]: The skill's instructions and metadata contain no prompt injection patterns, bypass instructions, or attempts to override system safety protocols.

scanpy